# Soluble cytokines enhance risk prediction across all stages of classical Hodgkin lymphoma

## TL;DR

This study shows that measuring certain soluble cytokines improves predicting outcomes for classical Hodgkin lymphoma patients at all disease stages.

## Contribution

The study introduces a new prognostic model using soluble cytokines that works across all stages of classical Hodgkin lymphoma.

## Key findings

- High levels of sIL-6 and sCD30 are linked to worse outcomes and advanced disease features in cHL.
- A new model combining age, extranodal disease, and cytokine levels outperforms existing risk models.
- Patients with elevated sCD30 and sIL-6 have significantly worse 5-year progression-free survival.

## Abstract

Classical Hodgkin lymphoma (cHL) is a heterogeneous malignancy with favorable outcomes, but accurate prognostic stratification remains challenging, particularly across all disease stages. Traditional risk models are focused on advanced stages and do not account for tumor microenvironment (TME) dynamics. Soluble cytokines reflecting TME activity may offer additional prognostic value. In our prospective multicenter study, we investigated the prognostic value of pretreatment plasma levels of soluble TARC, sCD30, sCD163, and sIL-6 in 162 newly diagnosed cHL patients and developed a model incorporating these biomarkers for risk prediction across all stages. Cytokine levels were measured using ELISA, and clinical characteristics, treatment responses, and outcomes were collected. The primary endpoint was 5-year progression-free survival (PFS). Elevated levels of sIL-6 and sCD30 were associated with higher disease stage, presence of B-symptoms, extranodal involvement, and inferior 5-year PFS. A novel prognostic model incorporating age, extranodal disease, and high sCD30/sIL-6 levels outperformed IPS-3 in predicting therapy failure. Patients with both elevated sCD30 and sIL-6 levels at diagnosis had significantly worse outcomes. Integrating soluble cytokine biomarkers, particularly sIL-6 and sCD30, into prognostic models enhances risk stratification across all stages of cHL and supports future efforts toward biomarker-driven, personalized treatment strategies.

The online version contains supplementary material available at 10.1186/s40364-025-00881-0.